Lee Merkhofer Consulting Priority Systems

Technical Terms Used in Project Portfolio Management (Continued)





























Q sort

A method for analyzing the opinions of a group based on a series of rankings (sorts) produced by the individual group members. The Q sort was originally developed within the field of psychology as a means for measuring correlations among the views expressed by different people. The letter Q was assigned to differentiate the method from another approach used in psychology, known as the R method, for analyzing correlations among various factors (e.g., height versus age). The Q sort, in contrast, is based on looking for correlations in the views of different subjects. A version of the Q sort is a popular collaborative approach for prioritizing projects.

The process for ranking projects via a Q sort involves a series of "rounds," with each round consisting of individual assessments followed by a group discussion. To prepare for a Q sort, a short description of each candidate project is written on a card. The resulting card deck is reproduced, and the copies are distributed to group members. A discussion is then held to improve the group's collective understanding of the projects.

Next, each member of the group individually ranks the projects by ordering his or her cards. Oftentimes, the ranking is done in steps. For example, initially, projects may be divided into two categories: higher priority and lower priority. Then the categories are decomposed to provide more discrimination, for example, sub dividing projects in each category into highest priority, medium priority, and lowest priority. The process continues until each individual has a complete, ordered ranking involving every project. A facilitator then tabulates the results and displays them to the group as a chart graph. Following the Delphi method for group assessments, results are reported anonymously, without associating people's names with their rankings. A period for discussion and debate follows, with participants arguing for various ranking positions for various projects. The process is then repeated in a second round, again with discussion following the anonymous reporting of individual priorities. By the third round, the facilitator usually attempts to lead the group to a consensus ranking of the projects.

The Q sort is popular, no doubt, because it is simple, easy to understand, and allows all participants equal influence in the decision-making process. Experience shows that it is a highly effective and efficient way of promoting consensus among participants. Its limitations are those associated with all methods that do not prioritize projects based on explicit criteria, analysis, and benefit-versus-cost logic. The process is not very transparent and does not generate documentation useful for explaining the reasoning underlying priorities—to outsiders, it may appear entirely political. Also, the Q sort depends on the participants having a complete and impartial understanding of each and every project, the needs the project serves, and the project's effectiveness. If there are a great many projects or if participants do not have an equally good understanding of all issues important to every project, basing priorities on the popular vote that underlies the Q sort is not likely to produce an optimal project portfolio.

quadratic programming

Similar to linear programming, except that the goal of the optimization is to maximize or minimize a quadratic function of the decision variables, for example:

ax12 + bx22 + cx1x2 ...

As with a linear program, there can be one or more linear constraints, for example,

Ax1 + Bx2N

There are many practical applications of quadratic programming. For example, modern portfolio theory identifies optimal investment portfolios by minimizing a quadratic function representing portfolio risk (the sum of the variances and covariances of the individual investments) subject to a linear constraint (a minimum expected return from the portfolio).

quality assurance (QA)

The methods that are designed and used by an organization to ensure that the activities it conducts and their results meet necessary quality standards.

qualitative risk analysis

An analysis that produces an assessment and characterization of risk in qualitative terms, such as providing a word description of the magnitude and potential consequences of the risk. Qualitative risk assessment may use numerical rating scales to produce a relative risk ranking. Popular qualitative risk assessment methods include scenario analysis, checklists, and a risk matrix. The advantage of qualitative risk assessment is that it relatively quick, easy, and inexpensive. Qualitative risk assessment is often used for the purpose of identifying larger risks that should be subjected to quantitative risk assessment. .

quantitative risk analysis (QRA)

An analysis that assesses and characterizes risk quantitatively in terms of the frequency or likelihood and magnitude of potential consequences. QRA involves creating a mathematical model of a project or process that explicitly includes uncertain variable that cannot be entirely controlled, and also decision variables that can be controlled. .The mathematical model can be simple or complex. The uncertainty over the model's uncertain variables is quantified either by using empirical data or by quantifying the judgments of knowledgeable experts. The analysis calculates the impact of the uncertain parameters and the decisions we make on outcomes that we care about -- such as profit and loss, investment returns, environmental consequences, and the like. QRA is required by regulations for certain projects and industries. Banks in particular are required by their regulators to identify and quantify their risks, often computing measures such as Value at Risk (VaR), and ensure that they have adequate capital to maintain solvency should the worst (or near-worst) outcomes occur.